psyc2071

Cognitivism

mind is a kind of unknown biological machine and scientists may propose theories about its structure

Computational metaphor

we can use the language of computing to build models attempting to explain mental processes

Ways of measuring cognitive processes

Accuracy of response, type of response, response time, neurological deficits, brain imaging, self-report

Neural networks

brain's information processing system

Universal computing machine

capable of computing any info

Types of universal computing machines

Turing machine, digital computers, Wang tiles, neural networks

Marr's levels of analysis

computational, algorithmic, implementational

Focused/selective attention

attending to one thing and ignoring other things

Divided attention

tracking multiple things

External attention

target of attention is something in the perceptual environment

Cross-modal attention

external attention; attending to multiple senses

Internal attention

target if attention is something you're thinking about

Active/endogenous attention

exercising top down (choosing) control over attention

Passive/exogenous attention

exercising bottom up (captured) attention where something in the environment calls attention to itself

Theories of auditory attention

early selection, late selection, attenuation

Cocktail party problem

lots of auditory sources, many signals, signals are jumbled. An attentional filter separates the sources and selects the target for processing

Early selection

pre-attentive low-level analysis picks up basic properties of different signals; attention selects a target for further processing; processing of unwanted signals stops and we are aware only of the results of pre-attentive analysis

semantic interference

Why not early selection?

semantic interference

semantic content of unattended source influences processing e.g. slowed reaction time

shadowing

subjects hear two audio channels over

headphones and attend to only one of them

Late selection

attention filters after meaning analysed; all signals/sources are processed however we aren't aware of this because our memory has limited capacity

Attenuation theory

every signal is processed only to the point where it can be separated from the target and then processing stops; some sources processed more than others

semantic interference (vs visual)

Negative priming is the visual equivalent of _

Serial search

visual search process that focuses on identifying a previously requested target surrounded by distractors possessing one or more common visual features with the target itself.

pop out effects

when target is defined by a specific feature it seems to "pop out"; set size makes no difference; type of feature doesn't matter

Parallel feature detection

feature maps (e.g. colour, orientation) analyse the visual fiel in parallel, hits on a particular map lead to fast detection of target item

slow serial search

needed when target is a combination of attributes, we need attention to bind features into objects

implications of slow serial search

binding of features happens late in the identification process and requires attention

Triesman's feature integration theory

feature extraction occurs automatically in parallel but object recognition requires feature binding

similarity helps us

form categories and make generalisations

similarity is

perceptual and conceptual phenomenon

Measuring similarity

confusability, reaction time, forced choice, likert scales

Universal law of generalisation

probability of generalising from one stimulus to another decreases exponentially as a function of dissimilarity (distance)

Problem with geometric approach

symmetry constraint - A is more like B than B is like A despite distance being equal

Featural similarity

assymetric knowledge e.g. know more about A than B, leads us to conclude certain features are distinctive to A, features common to A and B, features distinctive to B. Features in common increase similarity, features distinctive decrease similarity

Structure alignment

matching relations between comparison items. When 2 objects share a feature and it appears on the same spot in both items (match in place - MIP); when shared feature appears in a different location (match out of place - MOP)

similarity increases

MIPs and MOPS cause

MIPs

which are more effective in increasing similarity judgements - MIPs or MOPs?

stimulus transformation

short transformations: more similar, more confusable, slower reaction time (harder to distinguish); long transformations: less similar, less confusable, faster RT (easier to distinguish)

types of reasoning

deductive, inductive and informal are

Deductive

using facts to reach a logically certain conclusion

Inductive

using facts to reach a plausible conclusion

Syllogism

type of deductive reasoning; a conclusion is drawn from 2 given premises

Major premise

states a general rule

Minor premise

what part of argument states a specific fact

Valid argument

impossible for conclusion to be false if premises are true

types of valid arguments

modus ponens, modus tollens

modus ponens

affirming the antecedent; minor premises asserts that the antecedent of the major premise is true

modus tollens

denying the consequent; minor premise asserts that the consequent of the major premise is false

Invalid arguments

conclusion may be true but it doesn't necessarily follow from the premises

Types of invalid arguments

affirming the consequent: minor premise asserts that the consequent of the major premise is true; denying the antecedent: minor premises asserts that the antecedent of the major premise is false

antecedent related

what arguments are we good at?

consequent related

what arguments are we bad at?

assume affirmatory arguments correct

how do children evaluate arguments?

positive test bias

what bias do people show in Wason's selection task?

Inductive

relies on limited evidence to make a conclusion seem more plausible

premise-conclusion similarity

people are more likely to endorse an inductive argument when the premise and conclusion items are similar

premise diversity

people are more willing to endorse an inductive argument when the premises are dissimilar

premise monotonicity

people are more willing to make inductive generalisations when they have more examples

propositional fallacies

affirming the consequent, denying the antecedent

Informal

something wrong with content e.g. argument from ignorance, circular argument

Circular argument

assuming X is true in order to prove X is true

when you strongly accept the system as an explanation for a large body of facts

When are circular arguments (subjectively) strong?

Plausibility

low _ unlikely to be accepted; we rate circular arguments as more convincing when the alternative explanation is of lower _

Premise non-monotonicity

adding evidence weakens argument

Similarity

Why does premise non-monotonicity occur?

relevant story + relevant examples

Strong premise non-monotonicity

neutral story + relevant examples

slight premise non-monotonicity

neutral story + random examples

slight premise monotonicity

random story + random examples

strong premise monotonicity

procedural

type of memory is concerned with 'how' to do things; ability to perform skilled actions; unable to explicitly describe this

declarative

type of memory is concerned with 'what' knowledge; consists of episodic and semantic memory

episodic

memory for events

semantic

memory for facts/knowledge

concept

most basic unit of thought; abstract; learnt through exposure to examples of the concept

semantic features

concepts are represented in semantic memory through these; are so typical of the concept that they are virtually required if the object is going to be classified as belonging to that category

semantic network

hierarchical account of semantic memory; includes concept nodes and semantic features

sentence verification task

what task shows that some associations are made quicker than others?

probabilistic association

some concepts 100% associated with certain features and less strongly associated with other features

prototype model

concepts centred around a prototype which combines the most typical features of all members of that category; some items therefore have greater 'family resemblance' to the prototype than others

superordinate, basic, subordinate

Rosh's hierarchy of concepts

superordinate

broad category

basic concept

general item

subordinate

specific item

events

conjunctions of concepts

propositional networks

How do we represent events?

proposition

smallest idea that can be verified as true or false; includes a predicate and argument; abstract; language and modality free

time/context tag

how do we remember the specific location an event occurred? (re: propositional networks)

repeated exposures

how do propositional networks become stronger?

script

network of propositions

deviations

high recall of script _ because they are highly salient

ordinary/typical

poor recall of _ script parts

lexical

memory for concepts that we store in semantic memory

lexical processing

how do we store and access the words we know in lexical memory?

first part

most important part of a word

common words

in the lexical decision task, what type of words are accessed the fastest?

letter identity more important than letter position

a word can be read fairly well even if its middle letters have been transposed or moved about. what does this suggest?

semantic priming

words related to a prior prime are responded to faster

maintenance rehearsal

recycling of info; Miller's 7+/- 2 chunks